Southern Pacific Co. v. Arizona Ex Rel. Sullivan

325 U.S. 761 (1945)

SOUTHERN PACIFIC CO.
v.
ARIZONA EX REL. SULLIVAN, ATTORNEY GENERAL.

No. 56.

Supreme Court of United States.

Argued March 26, 27, 1945. Decided June 18, 1945. APPEAL FROM THE SUPREME COURT OF ARIZONA.

Messrs. Burton Mason and J. Carter Fort argued the cause, and Messrs. Cleon T. Knapp and C.W. Durbrow were with Mr. Mason on the brief, for appellant.

Messrs. Harold N. McLaughlin and Harold C. Heiss argued the cause, and Joe Conway, Attorney General of Arizona, Earl Anderson, Assistant Attorney General, and Mr. Charles L. Strouss were with Mr. McLaughlin on the brief, for appellee.

Mr. Robert L. Stern, with whom Solicitor General Fahy and Mrs. Carolyn R. Just were on the brief, for the United States, as amicus curiae, urging reversal.

Messrs. J. Carter Fort and Thomas Reed Powell filed a brief on behalf of the Association of American Railroads, as amicus curiae, urging reversal.

MR. CHIEF JUSTICE STONE delivered the opinion of the Court.

The Arizona Train Limit Law of May 16, 1912, Arizona Code Ann., 1939, § 69-119, makes it unlawful for any person or corporation to operate within the state a railroad train of more than fourteen passenger or seventy freight cars, and authorizes the state to recover a money penalty for each violation of the Act. The questions for decision are whether Congress has, by legislative enactment, restricted the power of the states to regulate the length of interstate trains as a safety measure and, if not, whether the statute contravenes the commerce clause of the Federal Constitution.

In 1940 the State of Arizona brought suit in the Arizona Superior Court against appellant, the Southern Pacific Company, to recover the statutory penalties for operating within the state two interstate trains, one a passenger train of more than fourteen cars, and one a freight train of more than seventy cars. Appellant answered, admitting the train operations, but defended on the ground that the statute offends against the commerce clause and the due process clause of the Fourteenth Amendment and conflicts with federal legislation. After an extended trial, without a jury, the court made detailed findings of fact on the basis of which it gave judgment for the railroad company. The Supreme Court of Arizona reversed and directed judgment for the state. 61 Ariz. 66, 145 P.2d 530. The case comes here on appeal under § 237 (a) of the Judicial Code, appellant raising by its assignments of error the questions presented here for decision.

The Supreme Court left undisturbed the findings of the trial court and made no new findings. It held that the power of the state to regulate the length of interstate trains had not been restricted by Congressional action. It sustained the Act as a safety measure to reduce the number of accidents attributed to the operation of trains of more than the statutory maximum length, enacted by the state legislature in the exercise of its "police power." This power the court held extended to the regulation of the operations of interstate commerce in the interests of local health, safety and well-being. It thought that a state statute, enacted in the exercise of the police power, and bearing some reasonable relation to the health, safety and well-being of the people of the state, of which the state legislature is the judge, was not to be judicially overturned, notwithstanding its admittedly adverse effect on the operation of interstate trains.

Purporting to act under § 1, paragraphs 10-17 of the Interstate Commerce Act, 24 Stat. 379 as amended (49 U.S.C. § 1 et seq.), the Interstate Commerce Commission, as of September 15, 1942, promulgated as an emergency measure Service Order No. 85, 7 Fed. Reg. 7258, suspending the operation of state train limit laws for the duration of the war, and denied an application to set aside the order. In the Matter of Service Order No. 85, 256 I.C.C. 523. Paragraph 15 of § 1 of the Interstate Commerce Act empowers the Commission, when it is "of opinion that shortage of equipment, congestion of traffic, or other emergency requiring immediate action exists in any section of the country," to make or suspend rules and practices "with respect to car service," which includes by paragraph 10 of § 1 "the use, control, supply, movement, distribution, exchange, interchange, and return" of locomotives and cars, and the "supply of trains." Paragraph 16 of § 1 provides that when a carrier is unable properly to transport the traffic offered, the Commission may make reasonable directions "with respect to the handling, routing, and movement of the traffic of such carrier and its distribution over other lines of roads." The authority of the Commission to make Order No. 85 is currently under attack in Johnston v. United States, Civil Action No. 1408, pending in the Western District of Oklahoma.

The Commission's order was not in effect in 1940 when the present suit was brought for violations of the state law in that year, and the Commission's order is inapplicable to the train operations here charged as violations. Hence the question here is not of the effect of the Commission's order, which we assume for purposes of decision to be valid, but whether the grant of power to the Commission operated to supersede the state act before the Commission's order. We are of opinion that, in the absence of administrative implementation by the Commission, § 1 does not of itself curtail state power to regulate train lengths. The provisions under which the Commission purported to act, phrased in broad and general language, do not in terms deal with that subject. We do not gain either from their words or from the legislative history any hint that Congress in enacting them intended, apart from Commission action, to supersede state laws regulating train lengths. We can hardly suppose that Congress, merely by conferring authority on the Commission to regulate car service in an "emergency," intended to restrict the exercise, otherwise lawful, of state power to regulate train lengths before the Commission finds an "emergency" to exist.

Congress, in enacting legislation within its constitutional authority over interstate commerce, will not be deemed to have intended to strike down a state statute designed to protect the health and safety of the public unless its purpose to do so is clearly manifested, Reid v. Colorado, 187 U.S. 137, 148; Missouri Pacific R. Co. v. Larabee Mills, 211 U.S. 612, 621, et seq.; Missouri, K. & T.R. Co. v. Harris, 234 U.S. 412, 418-419; Welch Co. v. New Hampshire, 306 U.S. 79, 85; Allen-Bradley Local v. Board, 315 U.S. 740, 749, or unless the state law, in terms or in its practical administration, conflicts with the Act of Congress, or plainly and palpably infringes its policy. Sinnot v. Davenport, 22 How. 227, 243; Missouri, K. & T.R. Co. v. Haber, 169 U.S. 613, 623; Savage v. Jones, 225 U.S. 501, 533; Carey v. South Dakota, 250 U.S. 118, 122; Atchison, T. & S.F.R. Co. v. Railroad Comm'n, 283 U.S. 380, 391; Townsend v. Yeomans, 301 U.S. 441, 454.

The contention, faintly urged, that the provisions of the Safety Appliance Act, 45 U.S.C. §§ 1 and 9, providing for brakes on trains, and of § 25 of Part I of the Interstate Commerce Act, 49 U.S.C. § 26 (b), permitting the Commission to order the installation of train stop and control devices, operate of their own force to exclude state regulation of train lengths, has even less support. Congress, although asked to do so,^[1] has declined to pass legislation specifically limiting trains to seventy cars. We are therefore brought to appellant's principal contention, that the state statute contravenes the commerce clause of the Federal Constitution.

Although the commerce clause conferred on the national government power to regulate commerce, its possession of the power does not exclude all state power of regulation. Ever since Willson v. Black-Bird Creek Marsh Co., 2 Pet. 245, and Cooley v. Board of Wardens, 12 How. 299, it has been recognized that, in the absence of conflicting legislation by Congress, there is a residuum of power in the state to make laws governing matters of local concern which nevertheless in some measure affect interstate commerce or even, to some extent, regulate it. Minnesota Rate Cases, 230 U.S. 352, 399-400; South Carolina Highway Dept. v. Barnwell Bros., 303 U.S. 177, 187, et seq.; California v. Thompson, 313 U.S. 109, 113-14 and cases cited; Parker v. Brown, 317 U.S. 341, 359-60. Thus the states may regulate matters which, because of their number and diversity, may never be adequately dealt with by Congress. Cooley v. Board of Wardens, supra, 319; South Carolina Highway Dept. v. Barnwell Bros., supra, 185; California v. Thompson, supra, 113; Duckworth v. Arkansas, 314 U.S. 390, 394; Parker v. Brown, supra, 362, 363. When the regulation of matters of local concern is local in character and effect, and its impact on the national commerce does not seriously interfere with its operation, and the consequent incentive to deal with them national is slight, such regulation has been generally held to be within state authority. South Carolina Highway Dept. v. Barnwell Bros., supra, 188 and cases cited; Lone Star Gas Co. v. Texas, 304 U.S. 224, 238; Milk Board v. Eisenberg Co., 306 U.S. 346, 351; Maurer v. Hamilton, 309 U.S. 598, 603; California v. Thompson, supra, 113-14 and cases cited.

But ever since Gibbons v. Ogden, 9 Wheat. 1, the states have not been deemed to have authority to impede substantially the free flow of commerce from state to state, or to regulate those phases of the national commerce which, because of the need of national uniformity, demand that their regulation, if any, be prescribed by a single authority.^[2]Cooley v. Board of Wardens, supra, 319; Leisy v. Hardin, 135 U.S. 100, 108, 109; Minnesota Rate Cases, supra, 399-400; Edwards v. California, 314 U.S. 160, 176. Whether or not this long-recognized distribution of power between the national and the state governments is predicated upon the implications of the commerce clause itself, Brown v. Maryland, 12 Wheat. 419, 447; Minnesota Rate Cases, supra, 399, 400; Pennsylvania v. West Virginia, 262 U.S. 553, 596; Baldwin v. Seelig, 294 U.S. 511, 522; South Carolina Highway Dept. v. Barnwell Bros., supra, 185, or upon the presumed intention of Congress, where Congress has not spoken, Welton v. Missouri, 91 U.S. 275, 282; Hall v. DeCuir, 95 U.S. 485, 490; Brown v. Houston, 114 U.S. 622, 631; Bowman v. Chicago & N.W.R. Co., 125 U.S. 465, 481-2; Leisy v. Hardin, supra, 109; In re Rahrer, 140 U.S. 545, 559-60; Brennan v. Titusville, 153 U.S. 289, 302; Covington & C. Bridge Co. v. Kentucky, 154 U.S. 204, 212; Graves v. New York ex rel. O'Keefe, 306 U.S. 466, 479, n., Dowling, Interstate Commerce and State Power, 27 Va. Law Rev. 1, the result is the same.

In the application of these principles some enactments may be found to be plainly within and others plainly without state power. But between these extremes lies the infinite variety of cases, in which regulation of local matters may also operate as a regulation of commerce, in which reconciliation of the conflicting claims of state and national power is to be attained only by some appraisal and accommodation of the competing demands of the state and national interests involved. Parker v. Brown, supra, 362; Terminal Railroad Assn. v. Brotherhood, 318 U.S. 1, 8; see DiSanto v. Pennsylvania, 273 U.S. 34, 44 (and compare California v. Thompson, supra); Illinois Gas Co. v. Public Service Co., 314 U.S. 498, 504-5.

For a hundred years it has been accepted constitutional doctrine that the commerce clause, without the aid of Congressional legislation, thus affords some protection from state legislation inimical to the national commerce, and that in such cases, where Congress has not acted, this Court, and not the state legislature, is under the commerce clause the final arbiter of the competing demands of state and national interests. Cooley v. Board of Wardens, supra; Kansas City Southern R. Co. v. Kaw Valley District, 233 U.S. 75, 79; South Covington R. Co. v. Covington, 235 U.S. 537, 546; Missouri, K. & T.R. Co. v. Texas, 245 U.S. 484, 488; St. Louis & S.F.R. Co. v. Public Service Comm'n, 254 U.S. 535, 537; Foster-Fountain Packing Co. v. Haydel, 278 U.S. 1, 10; Gwin, White & Prince v. Henneford, 305 U.S. 434, 441; McCarroll v. Dixie Lines, 309 U.S. 176.

Congress has undoubted power to redefine the distribution of power over interstate commerce. It may either permit the states to regulate the commerce in a manner which would otherwise not be permissible, In re Rahrer, supra, 561-62; Adams Express Co. v. Kentucky, 238 U.S. 190, 198; Rosenberger v. Pacific Express Co., 241 U.S. 48, 50, 51; Clark Distilling Co. v. Western Maryland R. Co., 242 U.S. 311, 325-6; Whitfield v. Ohio, 297 U.S. 431, 438-40; Kentucky Whip & Collar Co. v. Illinois Central R. Co., 299 U.S. 334, 350-51; Hooven & Allison Co. v. Evatt, 324 U.S. 652, 679, or exclude state regulation even of matters of peculiarly local concern which nevertheless affect interstate commerce. Addyston Pipe & Steel Co. v. United States, 175 U.S. 211, 230; Louisville & Nashville R. Co. v. Mottley, 219 U.S. 467; Houston, E. & W.T.R. Co. v. United States, 234 U.S. 342; American Express Co. v. Caldwell, 244 U.S. 617, 626; Illinois Central R. Co. v. Public Utilities Comm'n, 245 U.S. 493, 506; New York v. United States, 257 U.S. 591, 601; Louisiana Public Service Comm'n v. Texas & N.O.R. Co., 284 U.S. 125, 130; Pennsylvania R. Co. v. Illinois Brick Co., 297 U.S. 447, 459.

But in general Congress has left it to the courts to formulate the rules thus interpreting the commerce clause in its application, doubtless because it has appreciated the destructive consequences to the commerce of the nation if their protection were withdrawn, Gwin, White & Prince v. Henneford, supra, 441, and has been aware that in their application state laws will not be invalidated without the support of relevant factual material which will "afford a sure basis" for an informed judgment. Terminal Railroad Assn. v. Brotherhood, supra, 8; Southern R. Co. v. King, 217 U.S. 524. Meanwhile, Congress has accommodated its legislation, as have the states, to these rules as an established feature of our constitutional system. There has thus been left to the states wide scope for the regulation of matters of local state concern, even though it in some measure affects the commerce, provided it does not materially restrict the free flow of commerce across state lines, or interfere with it in matters with respect to which uniformity of regulation is of predominant national concern.

Hence the matters for ultimate determination here are the nature and extent of the burden which the state regulation of interstate trains, adopted as a safety measure, imposes on interstate commerce, and whether the relative weights of the state and national interests involved are such as to make inapplicable the rule, generally observed, that the free flow of interstate commerce and its freedom from local restraints in matters requiring uniformity of regulation are interests safeguarded by the commerce clause from state interference.

While this Court is not bound by the findings of the state court, and may determine for itself the facts of a case upon which an asserted federal right depends, Hooven & Allison Co. v. Evatt, supra, p. 659, and cases cited, the facts found by the state trial court showing the nature of the interstate commerce involved, and the effect upon it of the train limit law, are not seriously questioned. Its findings with respect to the need for and effect of the statute as a safety measure, although challenged in some particulars which we do not regard as material to our decision, are likewise supported by evidence. Taken together the findings supply an adequate basis for decision of the constitutional issue.

The findings show that the operation of long trains, that is trains of more than fourteen passenger and more than seventy freight cars, is standard practice over the main lines of the railroads of the United States, and that, if the length of trains is to be regulated at all, national uniformity in the regulation adopted, such as only Congress can prescribe, is practically indispensable to the operation of an efficient and economical national railway system. On many railroads passenger trains of more than fourteen cars and freight trains of more than seventy cars are operated, and on some systems freight trains are run ranging from one hundred and twenty-five to one hundred and sixty cars in length. Outside of Arizona, where the length of trains is not restricted, appellant runs a substantial proportion of long trains. In 1939 on its comparable route for through traffic through Utah and Nevada from 66 to 85% of its freight trains were over seventy cars in length and over 43% of its passenger trains included more than fourteen passenger cars.

In Arizona, approximately 93% of the freight traffic and 95% of the passenger traffic is interstate. Because of the Train Limit Law appellant is required to haul over 30% more trains in Arizona than would otherwise have been necessary. The record shows a definite relationship between operating costs and the length of trains, the increase in length resulting in a reduction of operating costs per car. The additional cost of operation of trains complying with the Train Limit Law in Arizona amounts for the two railroads traversing that state to about $1,000,000 a year. The reduction in train lengths also impedes efficient operation. More locomotives and more manpower are required; the necessary conversion and reconversion of train lengths at terminals and the delay caused by breaking up and remaking long trains upon entering and leaving the state in order to comply with the law, delays the traffic and diminishes its volume moved in a given time, especially when traffic is heavy.

To relieve the railroads of these burdens, during the war emergency only, the Interstate Commerce Commission, acting under § 1 of the Interstate Commerce Act, suspended the operation of the state law for the duration of the war by its order of September 15, 1942, to which we have referred. In support of the order the Commission declared: "It was designed to save manpower, motive power, engine-miles and train-miles; to avoid delay in the movement of trains; to increase the efficient use of locomotives and cars and to augment the available supply thereof; and to relieve congestion at terminals caused by setting out and picking up cars on each side of the train-limit law States." In the Matter of Service Order No. 85, 256 I.C.C. 523, 524. Appellant, because of its past compliance with the Arizona Train Limit Law, has been unable to avail itself fully of the benefits of the suspension order because some of its equipment and the length of its sidings in Arizona are not suitable for the operation of long trains. Engines capable of hauling long trains were not in service. It can engage in long train operations to the best advantage only by rebuilding its road to some extent and by changing or adding to its motive power equipment, which it desires to do in order to secure more efficient and economical operation of its trains.

The unchallenged findings leave no doubt that the Arizona Train Limit Law imposes a serious burden on the interstate commerce conducted by appellant. It materially impedes the movement of appellant's interstate trains through that state and interposes a substantial obstruction to the national policy proclaimed by Congress, to promote adequate, economical and efficient railway transportation service. Interstate Commerce Act, preceding § 1, 54 Stat. 899. Enforcement of the law in Arizona, while train lengths remain unregulated or are regulated by varying standards in other states, must inevitably result in an impairment of uniformity of efficient railroad operation because the railroads are subjected to regulation which is not uniform in its application. Compliance with a state statute limiting train lengths requires interstate trains of a length lawful in other states to be broken up and reconstituted as they enter each state according as it may impose varying limitations upon train lengths. The alternative is for the carrier to conform to the lowest train limit restriction of any of the states through which its trains pass, whose laws thus control the carriers' operations both within and without the regulating state.

Although the seventy car maximum for freight trains is the limitation which has been most commonly proposed, various bills introduced in the state legislatures provided for maximum freight train lengths of from fifty to one hundred and twenty-five cars, and maximum passenger train lengths of from ten to eighteen cars.^[3] With such laws in force in states which are interspersed with those having no limit on train lengths, the confusion and difficulty with which interstate operations would be burdened under the varied system of state regulation and the unsatisfied need for uniformity in such regulation, if any, are evident.^[4]

At present the seventy freight car laws are enforced only in Arizona and Oklahoma, with a fourteen car passenger car limit in Arizona. The record here shows that the enforcement of the Arizona statute results in freight trains being broken up and reformed at the California border and in New Mexico, some distance from the Arizona line. Frequently it is not feasible to operate a newly assembled train from the New Mexico yard nearest to Arizona, with the result that the Arizona limitation governs the flow of traffic as far east as El Paso, Texas. For similar reasons the Arizona law often controls the length of passenger trains all the way from Los Angeles to El Paso.^[5]

If one state may regulate train lengths, so may all the others, and they need not prescribe the same maximum limitation. The practical effect of such regulation is to control train operations beyond the boundaries of the state exacting it because of the necessity of breaking up and reassembling long trains at the nearest terminal points before entering and after leaving the regulating state. The serious impediment to the free flow of commerce by the local regulation of train lengths and the practical necessity that such regulation, if any, must be prescribed by a single body having a nation-wide authority are apparent.

The trial court found that the Arizona law had no reasonable relation to safety, and made train operation more dangerous. Examination of the evidence and the detailed findings makes it clear that this conclusion was rested on facts found which indicate that such increased danger of accident and personal injury as may result from the greater length of trains is more than offset by the increase in the number of accidents resulting from the larger number of trains when train lengths are reduced. In considering the effect of the statute as a safety measure, therefore, the factor of controlling significance for present purposes is not whether there is basis for the conclusion of the Arizona Supreme Court that the increase in length of trains beyond the statutory maximum has an adverse effect upon safety of operation. The decisive question is whether in the circumstances the total effect of the law as a safety measure in reducing accidents and casualties is so slight or problematical as not to outweigh the national interest in keeping interstate commerce free from interferences which seriously impede it and subject it to local regulation which does not have a uniform effect on the interstate train journey which it interrupts.

The principal source of danger of accident from increased length of trains is the resulting increase of "slack action" of the train. Slack action is the amount of free movement of one car before it transmits its motion to an adjoining coupled car. This free movement results from the fact that in railroad practice cars are loosely coupled, and the coupling is often combined with a shock-absorbing device, a "draft gear," which, under stress, substantially increases the free movement as the train is started or stopped. Loose coupling is necessary to enable the train to proceed freely around curves and is an aid in starting heavy trains, since the application of the locomotive power to the train operates on each car in the train successively, and the power is thus utilized to start only one car at a time.

The slack action between cars due to loose couplings varies from seven-eighths of an inch to one and one-eighth inches and, with the added free movement due to the use of draft gears, may be as high as six or seven inches between cars. The length of the train increases the slack since the slack action of a train is the total of the free movement between its several cars. The amount of slack action has some effect on the severity of the shock of train movements, and on freight trains sometimes results in injuries to operatives, which most frequently occur to occupants of the caboose. The amount and severity of slack action, however, are not wholly dependent upon the length of train, as they may be affected by the mode and conditions of operation as to grades, speed, and load. And accidents due to slack action also occur in the operation of short trains. On comparison of the number of slack action accidents in Arizona with those in Nevada, where the length of trains is now unregulated, the trial court found that with substantially the same amount of traffic in each state the number of accidents was relatively the same in long as in short train operations. While accidents from slack action do occur in the operation of passenger trains, it does not appear that they are more frequent or the resulting shocks more severe on long than on short passenger trains. Nor does it appear that slack action accidents occurring on passenger trains, whatever their length, are of sufficient severity to cause serious injury or damage.

As the trial court found, reduction of the length of trains also tends to increase the number of accidents because of the increase in the number of trains. The application of the Arizona law compelled appellant to operate 30.08%, or 4,304, more freight trains in 1938 than would otherwise have been necessary. And the record amply supports the trial court's conclusion that the frequency of accidents is closely related to the number of trains run. The number of accidents due to grade crossing collisions between trains and motor vehicles and pedestrians, and to collisions between trains, which are usually far more serious than those due to slack action, and accidents due to locomotive failures, in general vary with the number of trains.^[6] Increase in the number of trains results in more starts and stops, more "meets" and "passes," and more switching movements, all tending to increase the number of accidents not only to train operatives and other railroad employees, but to passengers and members of the public exposed to danger by train operations.

Railroad statistics introduced into the record tend to show that this is the result of the application of the Arizona Train Limit Law to appellant, both with respect to all railroad casualties within the state and those affecting only trainmen whom the train limit law is supposed to protect. The accident rate in Arizona is much higher than on comparable lines elsewhere, where there is no regulation of length of trains. The record lends support to the trial court's conclusion that the train length limitation increased rather than diminished the number of accidents. This is shown by comparison of appellant's operations in Arizona with those in Nevada,^[7] and by comparison of operations of appellant and of the Santa Fe Railroad in Arizona with those of the same roads in New Mexico,^[8] and by like comparison between appellant's operations in Arizona and operations throughout the country.^[9]

Upon an examination of the whole case the trial court found that "if short-train operation may or should result in any decrease in the number or severity of the ``slack' or ``slack-surge' type of accidents or casualties, such decrease is substantially more than offset by the increased number of accidents and casualties from other causes that follow the arbitrary limitation of freight trains to 70 cars . . . and passenger trains to 14 cars."

We think, as the trial court found, that the Arizona Train Limit Law, viewed as a safety measure, affords at most slight and dubious advantage, if any, over unregulated train lengths, because it results in an increase in the number of trains and train operations and the consequent increase in train accidents of a character generally more severe than those due to slack action. Its undoubted effect on the commerce is the regulation, without securing uniformity, of the length of trains operated in interstate commerce, which lack is itself a primary cause of preventing the free flow of commerce by delaying it and by substantially increasing its cost and impairing its efficiency. In these respects the case differs from those where a state, by regulatory measures affecting the commerce, has removed or reduced safety hazards without substantial interference with the interstate movement of trains. Such are measures abolishing the car stove, New York, N.H. & H.R. Co. v. New York, 165 U.S. 628; requiring locomotives to be supplied with electric headlights, Atlantic Coast Line R. Co. v. Georgia, 234 U.S. 280; providing for full train crews, Chicago, R.I. & P.R. Co. v. Arkansas, 219 U.S. 453; St. Louis & I.M.R. Co. v. Arkansas, 240 U.S. 518; Missouri Pacific R. Co. v. Norwood, 283 U.S. 249; and for the equipment of freight trains with cabooses, Terminal Railroad Assn. v. Brotherhood, supra.

The principle that, without controlling Congressional action, a state may not regulate interstate commerce so as substantially to affect its flow or deprive it of needed uniformity in its regulation is not to be avoided by "simply invoking the convenient apologetics of the police power," Kansas City Southern R. Co. v. Kaw Valley District, supra, 79; Buck v. Kuykendall, 267 U.S. 307, 315. In the Kaw Valley case the Court held that the state was without constitutional power to order a railroad to remove a railroad bridge over which its interstate trains passed, as a means of preventing floods in the district and of improving its drainage, because it was "not pretended that local welfare needs the removal of the defendants' bridges at the expense of the dominant requirements of commerce with other States, but merely that it would be helped by raising them." And in Seaboard Air Line R. Co. v. Blackwell, 244 U.S. 310, it was held that the interference with interstate rail transportation resulting from a state statute requiring as a safety measure that trains come almost to a stop at grade crossings, outweigh the local interest in safety, when it appeared that compliance increased the scheduled running time more than six hours in a distance of one hundred and twenty-three miles. Cf. Southern R. Co. v. King, supra, where the crossings were less numerous and the burden to interstate commerce was not shown to be heavy; and see Erb v. Morasch, 177 U.S. 584.

Similarly the commerce clause has been held to invalidate local "police power" enactments fixing the number of cars in an interstate train and the number of passengers to be carried in each car, South Covington R. Co. v. Covington, supra, 547; regulating the segregation of colored passengers in interstate trains, Hall v. DeCuir, supra, 488-9; requiring burdensome intrastate stops of interstate trains, Illinois Central R. Co. v. Illinois, 163 U.S. 142; Cleveland, C., C. & St. L.R. Co. v. Illinois, 177 U.S. 514; Mississippi Railroad Comm'n v. Illinois Central R. Co., 203 U.S. 335; Herndon v. Chicago, R.I. & P.R. Co., 218 U.S. 135; St. Louis-S.F.R. Co. v. Public Service Comm'n, 261 U.S. 369; requiring an interstate railroad to detour its through passenger trains for the benefit of a small city, St. Louis & S.F.R. Co. v. Public Service Comm'n, supra; interfering with interstate commerce by requiring interstate trains to leave on time, Missouri, K. & T.R. Co. v. Texas, 245 U.S. 484; regulating car distribution to interstate shippers, St. Louis S.W.R. Co. v. Arkansas, 217 U.S. 136; or establishing venue provisions requiring railroads to defend accident suits at points distant from the place of injury and the residence and activities of the parties, Davis v. Farmers Co-operative Co., 262 U.S. 312; Michigan Central R. Co. v. Mix, 278 U.S. 492; cf. Denver & R.G.W.R. Co. v. Terte, 284 U.S. 284; see also Buck v. Kuykendall, supra; Foster-Fountain Packing Co. v. Haydel, supra; Baldwin v. Seelig, supra, 524; South Carolina Highway Dept. v. Barnwell Bros., supra, 184-5 n., and cases cited.

More recently in Kelly v. Washington, 302 U.S. 1, 15, we have pointed out that when a state goes beyond safety measures which are permissible because only local in their effect upon interstate commerce, and "attempts to impose particular standards as to structure, design, equipment and operation [of vessels plying interstate] which in the judgment of its authorities may be desirable but pass beyond what is plainly essential to safety and seaworthiness, the State will encounter the principle that such requirements, if imposed at all, must be through the action of Congress which can establish a uniform rule. Whether the State in a particular matter goes too far must be left to be determined when the precise question arises."

Here we conclude that the state does go too far. Its regulation of train lengths, admittedly obstructive to interstate train operation, and having a seriously adverse effect on transportation efficiency and economy, passes beyond what is plainly essential for safety since it does not appear that it will lessen rather than increase the danger of accident. Its attempted regulation of the operation of interstate trains cannot establish nation-wide control such as is essential to the maintenance of an efficient transportation system, which Congress alone can prescribe. The state interest cannot be preserved at the expense of the national interest by an enactment which regulates interstate train lengths without securing such control, which is a matter of national concern. To this the interest of the state here asserted is subordinate.

Appellees especially rely on the full train crew cases, Chicago, R.I. & P.R. Co. v. Arkansas, supra; St. Louis & I.M.R. Co. v. Arkansas, supra; Missouri Pacific R. Co. v. Norwood, supra, and also on South Carolina Highway Dept. v. Barnwell Bros., supra, as supporting the state's authority to regulate the length of interstate trains. While the full train crew laws undoubtedly placed an added financial burden on the railroads in order to serve a local interest, they did not obstruct interstate transportation or seriously impede it. They had no effects outside the state beyond those of picking up and setting down the extra employees at the state boundaries; they involved no wasted use of facilities or serious impairment of transportation efficiency, which are among the factors of controlling weight here. In sustaining those laws the Court considered the restriction a minimal burden on the commerce comparable to the law requiring the licensing of engineers as a safeguard against those of reckless and intemperate habits, sustained in Smith v. Alabama, 124 U.S. 465, or those afflicted with color blindness, upheld in Nashville, C. & St. L.R. Co. v. Alabama, 128 U.S. 96, and other similar regulations. New York, N.H. & H.R. Co. v. New York, supra; Atlantic Coast Line R. Co. v. Georgia, supra; cf. County of Mobile v. Kimball, 102 U.S. 691.

South Carolina Highway Dept. v. Barnwell Bros., supra, was concerned with the power of the state to regulate the weight and width of motor cars passing interstate over its highways, a legislative field over which the state has a far more extensive control than over interstate railroads. In that case, and in Maurer v. Hamilton, supra, we were at pains to point out that there are few subjects of state regulation affecting interstate commerce which are so peculiarly of local concern as is the use of the state's highways. Unlike the railroads local highways are built, owned and maintained by the state or its municipal subdivisions. The state is responsible for their safe and economical administration. Regulations affecting the safety of their use must be applied alike to intrastate and interstate traffic. The fact that they affect alike shippers in interstate and intrastate commerce in great numbers, within as well as without the state, is a safeguard against regulatory abuses. Their regulation is akin to quarantine measures, game laws, and like local regulations of rivers, harbors, piers, and docks, with respect to which the state has exceptional scope for the exercise of its regulatory power, and which, Congress not acting, have been sustained even though they materially interfere with interstate commerce (303 U.S. at 187-188 and cases cited).

The contrast between the present regulation and the full train crew laws in point of their effects on the commerce, and the like contrast with the highway safety regulations, in point of the nature of the subject of regulation and the state's interest in it, illustrate and emphasize the considerations which enter into a determination of the relative weights of state and national interests where state regulation affecting interstate commerce is attempted. Here examination of all the relevant factors makes it plain that the state interest is outweighed by the interest of the nation in an adequate, economical and efficient railway transportation service, which must prevail.

Reversed.

MR. JUSTICE RUTLEDGE concurs in the result.

MR. JUSTICE BLACK, dissenting.

In Hennington v. Georgia, 163 U.S. 299, 304, a case which involved the power of a state to regulate interstate traffic, this Court said, "The whole theory of our government, federal and state, is hostile to the idea that questions of legislative authority may depend . . . upon opinions of judges as to the wisdom or want of wisdom in the enactment of laws under powers clearly conferred upon the legislature." What the Court decides today is that it is unwise governmental policy to regulate the length of trains. I am therefore constrained to note my dissent.

For more than a quarter of a century, railroads and their employees have engaged in controversies over the relative virtues and dangers of long trains. Railroads have argued that they could carry goods and passengers cheaper in long trains than in short trains. They have also argued that while the danger of personal injury to their employees might in some respects be greater on account of the operation of long trains, this danger was more than offset by an increased number of accidents from other causes brought about by the operation of a much larger number of short trains. These arguments have been, and are now, vigorously denied. While there are others, the chief causes assigned for the belief that long trains unnecessarily jeopardize the lives and limbs of railroad employees relate to "slack action." Cars coupled together retain a certain free play of movement, ranging between 1 1/2 inches and 1 foot, and this is called "slack action." Train brakes do not ordinarily apply or release simultaneously on all cars. This frequently results in a severe shock or jar to cars, particularly those in the rear of a train. It has always been the position of the employees that the dangers from "slack action" correspond to and are proportionate with the length of the train. The argument that "slack movements" are more dangerous in long trains than in short trains seems never to have been denied. The railroads have answered it by what is in effect a plea of confession and avoidance. They say that the added cost of running short trains places an unconstitutional burden on interstate commerce. Their second answer is that the operation of short trains requires the use of more separate train units; that a certain number of accidents resulting in injury are inherent in the operation of each unit, injuries which may be inflicted either on employees or on the public; consequently, they have asserted that it is not in the public interest to prohibit the operation of long trains.

In 1912, the year Arizona became a state, its legislature adopted and referred to the people several safety measures concerning the operation of railroads. One of these required railroads to install electric headlights, a power which the state had under this Court's opinion in Atlantic Coast Line R. Co. v. Georgia, 234 U.S. 280. Another Arizona safety statute submitted at the same time required certain tests and service before a person could act as an engineer or train conductor, and thereby exercised a state power similar to that which this Court upheld in Nashville, C. & St. L.R. Co. v. Alabama, 128 U.S. 96. The third safety statute which the Arizona legislature submitted to the electorate, and which was adopted by it, is the train limitation statute now under consideration. By its enactment the legislature and the people adopted the viewpoint that long trains were more dangerous than short trains, and limited the operation of train units to 14 cars for passenger and 70 cars for freight. This same question was considered in other states, and some of them, over the vigorous protests of railroads, adopted laws similar to the Arizona statute.^[1]

This controversy between the railroads and their employees, which was nation-wide, was carried to Congress. Extensive hearings took place. The employees' position was urged by members of the various Brotherhoods. The railroads' viewpoint was presented through representatives of their National Association. In 1937, the Senate Interstate Commerce Committee after its own exhaustive hearings unanimously recommended that trains be limited to 70 cars as a safety measure.^[2] The Committee in its Report reviewed the evidence and specifically referred to the large and increasing number of injuries and deaths suffered by railroad employees; it concluded that the admitted danger from slack movement was greatly intensified by the operation of long that short trains reduce this danger; that the added cost of short trains to the railroad was no justification for jeopardizing the safety of railroad employees; and that the legislation would provide a greater degree of safety for persons and property, increase protection for railway employees and the public, and improve transportation services for shippers and consumers. The Senate passed the bill^[3] but the House Committee failed to report it out.

During the hearings on that measure, frequent references were made to the Arizona statute. It is significant, however, that American railroads never once asked Congress to exercise its unquestioned power to enact uniform legislation on that subject, and thereby invalidate the Arizona law. That which for some unexplained reason they did not ask Congress to do when it had the very subject of train length limitations under consideration, they shortly thereafter asked an Arizona state court to do.

In the state court a rather extraordinary "trial" took place. Charged with violating the law, the railroad admitted the charge. It alleged that the law was unconstitutional, however, and sought a trial of facts on that issue. The essence of its charge of unconstitutionality rested on one of these two grounds: (1) the legislature and people of Arizona erred in 1912 in determining that the running of long trains was dangerous; or (2) railroad conditions had so improved since 1912 that previous dangers did not exist to the same extent, and that the statute should be stricken down either because it cast an undue burden on interstate commerce by reason of the added cost, or because the changed conditions had rendered the Act "arbitrary and unreasonable." Thus, the issue which the court "tried" was not whether the railroad was guilty of violating the law, but whether the law was unconstitutional either because the legislature had been guilty of misjudging the facts concerning the degree of the danger of long trains, or because the 1912 conditions of danger no longer existed.

Before the state trial court finally determined that the dangers found by the legislature in 1912 no longer existed, it heard evidence over a period of 5 1/2 months which appears in about 3,000 pages of the printed record before us. It then adopted findings of fact submitted to it by the railroad, which cover 148 printed pages, and conclusions of law which cover 5 pages. We can best understand the nature of this "trial" by analogizing the same procedure to a defendant charged with violating a state or national safety appliance act, where the defendant comes into court and admits violation of the act. In such cases, the ordinary procedure would be for the court to pass upon the constitutionality of the act, and either discharge or convict the defendants. The procedure here, however, would justify quite a different trial method. Under it, a defendant is permitted to offer voluminous evidence to show that a legislative body has erroneously resolved disputed facts in finding a danger great enough to justify the passage of the law. This new pattern of trial procedure makes it necessary for a judge to hear all the evidence offered as to why a legislature passed a law and to make findings of fact as to the validity of those reasons. If under today's ruling a court does make findings, as to a danger contrary to the findings of the legislature, and the evidence heard "lends support" to those findings, a court can then invalidate the law. In this respect, the Arizona County Court acted, and this Court today is acting, as a "super-legislature."^[4]

Even if this method of invalidating legislative acts is a correct one, I still think that the "findings" of the state court do not authorize today's decision. That court did not find that there is no unusual danger from slack movements in long trains. It did decide on disputed evidence that the long train "slack movement" dangers were more than offset by prospective dangers as a result of running a larger number of short trains, since many people might be hurt at grade crossings. There was undoubtedly some evidence before the state court from which it could have reached such a conclusion. There was undoubtedly as much evidence before it which would have justified a different conclusion.

Under those circumstances, the determination of whether it is in the interest of society for the length of trains to be governmentally regulated is a matter of public policy. Someone must fix that policy — either the Congress, or the state, or the courts. A century and a half of constitutional history and government admonishes this Court to leave that choice to the elected legislative representatives of the people themselves, where it properly belongs both on democratic principles and the requirements of efficient government.

I think that legislatures, to the exclusion of courts, have the constitutional power to enact laws limiting train lengths, for the purpose of reducing injuries brought about by "slack movements." Their power is not less because a requirement of short trains might increase grade crossing accidents. This latter fact raises an entirely different element of danger which is itself subject to legislative regulation. For legislatures may, if necessary, require railroads to take appropriate steps to reduce the likelihood of injuries at grade crossings. Denver & R.G.R. Co. v. Denver, 250 U.S. 241. And the fact that grade-crossing improvements may be expensive is no sufficient reason to say that an unconstitutional "burden" is put upon a railroad even though it be an interstate road. Erie R. Co. v. Public Utility Commissioners, 254 U.S. 394, 408-411.

The Supreme Court of Arizona did not discuss the County Court's so-called findings of fact. It properly designated the Arizona statute as a safety measure, and finding that it bore a reasonable relation to its purpose declined to review the judgment of the legislature as to the necessity for the passage of the act. In so doing it was well fortified by a long line of decisions of this Court. Today's decision marks an abrupt departure from that line of cases.

There have been many sharp divisions of this Court concerning its authority, in the absence of congressional enactment, to invalidate state laws as violating the Commerce Clause. See e. g., Adams Manufacturing Co. v. Storen, 304 U.S. 307; Gwin, White & Prince v. Henneford, 305 U.S. 434; McCarroll v. Dixie Greyhound Lines, 309 U.S. 176. That discussion need not be renewed here, because even the broadest exponents of judicial power in this field have not heretofore expressed doubt as to a state's power, absent a paramount congressional declaration, to regulate interstate trains in the interest of safety. For as early as 1913, this Court, speaking through Mr. Justice Hughes, later Chief Justice, referred to "the settled principle that, in the absence of legislation by Congress, the states are not denied the exercise of their power to secure safety in the physical operation of railroad trains within their territory, even though such trains are used in interstate commerce. That has been the law since the beginning of railroad transportation." Atlantic Coast Line R. Co. v. Georgia, 234 U.S. 280, 291. Until today, the oft-repeated principles of that case have never been repudiated in whole or in part.

But, it is said today, the principle there announced does not apply because if one state applies a regulation of its own to interstate trains, "uniformity" in regulation or rather non-regulation, is destroyed. Justice Hughes speaking for the Court in the Atlantic Coast Line case made short shrift of that same argument. He there referred to the contention that "if state requirements conflict, it will be necessary to carry additional apparatus and to make various adjustments at state lines which would delay and inconvenience interstate traffic." In answer to this argument he reiterated a former declaration of this Court in New York, N.H. & H.R. Co. v. New York, 165 U.S. 628, on this subject, and added that "If there is a conflict in such local regulations, by which interstate commerce may be inconvenienced — if there appears to be need of standardization of safety appliances and of providing rules of operation which will govern the entire interstate road irrespective of state boundaries — there is a simple remedy; and it cannot be assumed that it will not be readily applied if there be real occasion for it. That remedy does not rest in a denial to the state, in the absence of conflicting federal action, of its power to protect life and property within its borders, but it does lie in the exercise of the paramount authority of Congress in its control of interstate commerce to establish such regulations as in its judgment may be deemed appropriate and sufficient. Congress, when it pleases, may give the rule and make the standard to be observed on the interstate highway." p. 292.

That same statement has in substance been made in many other decisions of this Court, a number of which are cited in the Atlantic Coast Line case, and all of them are today swept into the discard. In no one of all these previous cases was it more appropriate than here to call attention to the fact that Congress could when it pleased establish a uniform rule as to the length of trains. Congress knew about the Arizona law. It is common knowledge that the Interstate Commerce Committees of the House and the Senate keep in close and intimate touch with the affairs of railroads and other national means of transportation. Every year brings forth new legislation which goes through those Committees, much of it relating to safety. The attention of the members of Congress and of the Senate have been focused on the particular problem of the length of railroad trains. We cannot assume that they were ignorant of the commonly known fact that a long train might be more dangerous in some territories and on some particular types of railroad. The history of congressional consideration of this problem leaves little if any room to doubt that the choice of Congress to leave the state free in this field was a deliberate choice, which was taken with a full knowledge of the complexities of the problems and the probable need for diverse regulations in different localities. I am therefore compelled to reach the conclusion that today's decision is the result of the belief of a majority of this Court that both the legislature of Arizona and the Congress made wrong policy decisions in permitting a law to stand which limits the length of railroad trains. I should at least give the Arizona statute the benefit of the same rule which this Court said should be applied in connection with state legislation under attack for violating the Fourteenth Amendment, that is, that legislative bodies have "a wide range of legislative discretion, . . . and their conclusions respecting the wisdom of their legislative acts are not reviewable by the courts." Arizona Employers' Liability Cases, 250 U.S. 400, 419.

When we finally get down to the gist of what the Court today actually decides, it is this: Even though more railroad employees will be injured by "slack action" movements on long trains than on short trains, there must be no regulation of this danger in the absence of "uniform regulations." That means that no one can legislate against this danger except the Congress; and even though the Congress is perfectly content to leave the matter to the different state legislatures, this Court, on the ground of "lack of uniformity," will require it to make an express avowal of that fact before it will permit a state to guard against that admitted danger.

We are not left in doubt as to why, as against the potential peril of injuries to employees, the Court tips the scales on the side of "uniformity." For the evil it finds in a lack of uniformity is that it (1) delays interstate commerce, (2) increases its cost and (3) impairs its efficiency. All three of these boil down to the same thing, and that is that running shorter trains would increase the cost of railroad operations. The "burden" on commerce reduces itself to mere cost because there was no finding, and no evidence to support a finding, that by the expenditure of sufficient sums of money, the railroads could not enable themselves to carry goods and passengers just as quickly and efficiently with short trains as with long trains. Thus the conclusion that a requirement for long trains will "burden interstate commerce" is a mere euphemism for the statement that a requirement for long trains will increase the cost of railroad operations.

In the report of the Senate Committee, supra, attention was called to the fact that in 1935, 6,351 railroad employees were injured while on duty, with a resulting loss of more than 200,000 working days, and that injuries to trainmen and enginemen increased more than 29% in 1936.^[5] Nevertheless, the Court's action in requiring that money costs outweigh human values is sought to be buttressed by a reference to the express policy of Congress to promote an "economical national railroad system." I cannot believe that if Congress had defined what it meant by "economical," it would have required money to be saved at the expense of the personal safety of railway employees. Its whole history for the past 25 years belies such an interpretation of its language. Judicial opinions rather than legislative enactments have tended to emphasize costs. See Tiller v. Atlantic Coast Line R. Co., supra, 58-60. A different congressional attitude has been shown by the passage of numerous safety appliance provisions, a federal employees' compensation act, abolition of the judicially created doctrine of assumption of risk and contributory negligence, and various other types of legislation. Unfortunately, the record shows, as pointed out in the Tiller case, that the courts have by narrow and restricted interpretation too frequently reduced the full scope of protection which Congress intended to provide.

This record in its entirety leaves me with no doubt whatever that many employees have been seriously injured and killed in the past, and that many more are likely to be so in the future, because of "slack movement" in trains. Everyday knowledge as well as direct evidence presented at the various hearings, substantiates the report of the Senate Committee that the danger from slack movement is greater in long trains than in short trains. It may be that offsetting dangers are possible in the operation of short trains. The balancing of these probabilities, however, is not in my judgment a matter for judicial determination, but one which calls for legislative consideration. Representatives elected by the people to make their laws, rather than judges appointed to interpret those laws, can best determine the policies which govern the people. That at least is the basic principle on which our democratic society rests. I would affirm the judgment of the Supreme Court of Arizona.

MR. JUSTICE DOUGLAS, dissenting.

I have expressed my doubts whether the courts should intervene in situations like the present and strike down state legislation on the grounds that it burdens interstate commerce. McCarroll v. Dixie Greyhound Lines, 309 U.S. 176, 183-189. My view has been that the courts should intervene only where the state legislation discriminated against interstate commerce or was out of harmony with laws which Congress had enacted. p. 184. It seems to me particularly appropriate that that course be followed here. For Congress has given the Interstate Commerce Commission broad powers of regulation over interstate carriers. The Commission is the national agency which has been entrusted with the task of promoting a safe, adequate, efficient, and economical transportation service. It is the expert on this subject. It is in a position to police the field. And if its powers prove inadequate for the task, Congress, which has paramount authority in this field, can implement them.

But the Court has not taken that view. As a result the question presented is whether the total effect of Arizona's train-limit as a safety measure is so slight as not to outweigh the national interest in keeping interstate commerce free from interferences which seriously impede or burden it. The voluminous evidence has been reviewed in the opinion of the Court and in the dissenting opinion of MR. JUSTICE BLACK. If I sat as a member of the Interstate Commerce Commission or of a legislative committee to decide whether Arizona's train-limit law should be superseded by a federal regulation, the question would not be free from doubt for me. If we had before us the ruling of the Interstate Commerce Commission (In the Matter of Service Order No. 85, 256 I.C.C. 523, 534) that Arizona's train-limit law infringes "the national interest in maintaining the free flow of commerce under the present emergency war conditions," I would accept its expert appraisal of the facts, assuming it had the authority to act. But that order is not before us. And the present case deals with a period of time which antedates the war emergency. Moreover, we are dealing here with state legislation in the field of safety where the propriety of local regulation has long been recognized. See Atlantic Coast Line R. Co. v. Georgia, 234 U.S. 280, 291, and cases collected in California v. Thompson, 313 U.S. 109, 113-114. Whether the question arises under the Commerce Clause or the Fourteenth Amendment, I think the legislation is entitled to a presumption of validity. If a State passed a law prohibiting the hauling of more than one freight car at a time, we would have a situation comparable in effect to a state law requiring all railroads within its borders to operate on narrow gauge tracks. The question is one of degree and calls for a close appraisal of the facts.^[1] I am not persuaded that the evidence adduced by the railroads overcomes the presumption of validity to which this train-limit law is entitled. For the reasons stated by MR. JUSTICE BLACK, Arizona's train-limit law should stand as an allowable regulation enacted to protect the lives and limbs of the men who operate the trains.

NOTES

[1] See Senate Report No. 416, 75th Cong., 1st Sess.; 81 Cong. Rec. 7596; and Hearings before House Committee on Interstate and Foreign Commerce, 75th Cong., 3d Sess., S. 69, Train Lengths.

[2] In applying this rule the Court has often recognized that to the extent that the burden of state regulation falls on interests outside the state, it is unlikely to be alleviated by the operation of those political restraints normally exerted when interests within the state are affected. Cooley v. Board of Wardens, supra, 315; Gilman v. Philadelphia, 3 Wall. 713, 731; Escanaba Co. v. Chicago, 107 U.S. 678, 683; Robbins v. Shelby County Taxing Dist., 120 U.S. 489, 499; Lake Shore & M.S.R. Co. v. Ohio, 173 U.S. 285, 294; South Carolina Highway Dept. v. Barnwell Bros. supra, 185, n.; McGoldrick v. Berwind-White Co., 309 U.S. 33, 46, n.; cf. McCulloch v. Maryland, 4 Wheat. 316, 428; Pound v. Turck, 95 U.S. 459, 464; Gloucester Ferry Co. v. Pennsylvania, 114 U.S. 196, 205; Helvering v. Gerhardt, 304 U.S. 405, 412.

[3] One hundred sixty-four bills limiting train lengths have been introduced in state legislatures since 1920, of which only three were passed, in Nevada, Louisiana and Oklahoma. The Nevada and Louisiana laws were held unconstitutional and never enforced. Southern Pacific Co. v. Mashburn, 18 F. Supp. 393; Texas & N.O.R. Co. v. Martin (No. 428 — Equity, E.D. of La. 1936), unreported. The Arizona law, passed in 1912, was held unconstitutional in Atchison, T. & S.F.R. Co. v. La Prade, 2 F. Supp. 855, reversed on other grounds, 289 U.S. 444.

[4] Had these bills been passed a freight train running over established routes (from Virginia to Michigan for example) would normally proceed through states with a seventy-five car maximum (Virginia), a one hundred and twenty-five car maximum (West Virginia), a three thousand foot maximum (Ohio), and a seventy car limit (Michigan). A train from Arkansas to Wisconsin might be subjected to a fifty car maximum (Arkansas), one-half mile (Mississippi), three thousand feet (Iowa), one and a half miles (Minnesota), and thirty-three hundred feet (Wisconsin). A train running from Nebraska to California might be subject to a sixty, seventy-five or eighty-five maximum in Nebraska, to a limit fixed by commission in Kansas, to a sixty-five car limit in Colorado, to a seventy-five car limit in New Mexico, to a seventy car limit in Arizona, and to a seventy-four car limit in California. A passenger train might be limited to fourteen cars in New Jersey, ten in Pennsylvania and eighteen in West Virginia.

[5] In Oklahoma three lines running from Chicago or Kansas City west pass through Oklahoma for distances of sixty, one hundred and seventeen and one hundred and forty-three miles. Since no other state through which the traffic passes (except Arizona) restricts train lengths in any way, the effect of the Oklahoma law is to require through trains to be broken up for the short distances they pass through that state.

[6] The record shows that in 1939 the number of slack accident casualties in the United States, 399, was only 6% of the number of train and train service casualties to railroad employees, 6,713. In that year three of the 399 slack accident casualties were fatal, whereas the average number of grade crossing casualties per year from 1935 to 1939 was 5,718. And in 1939, 1,398 persons were killed and 3,999 were injured in highway, grade crossing accidents. I.C.C., Bureau of Statistics, Accident Bulletin, No. 108, pp. 22-23.

[7] With passenger traffic in Nevada 78% as heavy as in Arizona, from 1923 to 1938 two hundred and thirty-nine casualties were caused to persons by passenger trains in Arizona and one hundred and nine in Nevada. Between 1923 and 1939 five persons in Nevada and fourteen in Arizona were injured by sudden stops or jerks on passenger trains.

[8] Casualties to employees, occurring in freight train operations in New Mexico, have been substantially less in both number and frequency than in Arizona. From 1930 to 1940 there were one hundred and twenty-nine casualties to all classes of employees in New Mexico at the rate of 7.97 per million train miles, 12.84 per hundred million car miles. In Arizona there were two hundred and fifty-one casualties to employees, at the rate of 10.03 per million train miles, and 18.10 per hundred million car miles.

[9] On a national basis the findings show that while the national accident rate per hundred million car miles for all railroad employees and for trainmen decreased 70% to 66% respectively between 1923-1928 and 1935-1940, the rate for the Southern Pacific in Arizona declined 52.3% and 53.3%. Appellant's rate in Nevada decreased 71.1% and 69.1%.

[1] A resume of these laws and their reception by the courts is set out in the opinion of the Supreme Court of Arizona in this case, 61 Ariz. 66, 145 P.2d 530.

[2] Senate Report No. 416, 75th Cong., 1st Sess.

[3] 81 Cong. Rec. 7596. The record does not show any dissenting votes cast against the bill. The debate on the measure appears at pp. 7564-7595.

[4] The Court today invalidates the Arizona law in accordance with the identical "super-legislature" method (so designated by Justices Brandeis and Holmes) used by the majority to invalidate a Nebraska statute regulating the weights of loaves of bread. Burns Baking Co. v. Bryan, 264 U.S. 504, 534. For here, as there, this Court has overruled a state legislature's finding that an evil existed, and that the state law would not impose an unconstitutional "burden" upon those regulated. The dissent in the Burns case said:

"To decide, as a fact, that the prohibition of excess weights ``is not necessary for the protection of the purchasers against imposition and fraud by short weights'; that it ``is not calculated to effectuate that purpose'; and that it ``subjects bakers and sellers of bread' to heavy burdens, is, in my opinion, an exercise of the powers of a super-legislature — not the performance of the constitutional function of judicial review."

That decision rested on the Due Process Clause while today's decision rests on the Commerce Clause. But that difference does not make inapplicable here the principles invoked by the dissenters in the Burns case.

The use of the "super-legislature" technique has been repeated to strike down other statutes. See e.g., Chicago, M. & St. P.R. Co. v. Wisconsin, 238 U.S. 491, 499; Weaver v. Palmer Bros. Co. 270 U.S. 402, dissent at 415. See also dissents in Schlesinger v. Wisconsin, 270 U.S. 230, 241, 242; New State Ice Co. v. Liebmann, 285 U.S. 262, 284-285. For a case in which this Court declined to review the "economics or the facts" behind a legislative enactment, see Central Lumber Co. v. South Dakota, 226 U.S. 157, 161; cf. Standard Oil Co. v. Marysville, 279 U.S. 582, 586. See also Powell v. Pennsylvania, 127 U.S. 678, 686; dissenting opinion, Polk Co. v. Glover, 305 U.S. 5, 10-19.

[5] These figures appear to be considerably less than those later reported. See Tiller v. Atlantic Coast Line R. Co., 318 U.S. 54, 59, note 4.

[1] See Bikle, Judicial Determination of Questions of Fact Affecting The Constitutional Validity of Legislative Action, 38 Harv. L. Rev. 6.